Special Articles

Recommendations for end-of-life care in the intensive care

unit: A consensus statement by the American College of
Critical Care Medicine
Robert D. Truog, MD, MA; Margaret L. Campbell, PhD, RN, FAAN; J. Randall Curtis, MD, MPH;
Curtis E. Haas, PharmD, FCCP; John M. Luce, MD; Gordon D. Rubenfeld, MD, MSc;
Cynda Hylton Rushton, PhD, RN, FAAN; David C. Kaufman, MD

Background: These recommendations have been developed to
improve the care of intensive care unit (ICU) patients during the
dying process. The recommendations build on those published in
2003 and highlight recent developments in the field from a U.S.
perspective. They do not use an evidence grading system because
most of the recommendations are based on ethical and legal
principles that are not derived from empirically based evidence.
Principal Findings: Family-centered care, which emphasizes
the importance of the social structure within which patients are
embedded, has emerged as a comprehensive ideal for managing
end-of-life care in the ICU. ICU clinicians should be competent in
all aspects of this care, including the practical and ethical aspects
of withdrawing different modalities of life-sustaining treatment
and the use of sedatives, analgesics, and nonpharmacologic
approaches to easing the suffering of the dying process. Several
key ethical concepts play a foundational role in guiding end-of-
life care, including the distinctions between withholding and
withdrawing treatments, between actions of killing and allowing
to die, and between consequences that are intended vs. those that
are merely foreseen (the doctrine of double effect). Improved
communication with the family has been shown to improve pa-
tient care and family outcomes. Other knowledge unique to end-
of-life care includes principles for notifying families of a patient’s
death and compassionate approaches to discussing options for
organ donation. End-of-life care continues even after the death of
the patient, and ICUs should consider developing comprehensive
bereavement programs to support both families and the needs of
the clinical staff. Finally, a comprehensive agenda for improving
end-of-life care in the ICU has been developed to guide research,
quality improvement efforts, and educational curricula.
Conclusions: End-of-life care is emerging as a comprehensive
area of expertise in the ICU and demands the same high level of
knowledge and competence as all other areas of ICU practice.
(Crit Care Med 2008; 36:953–963)
KEY WORDS: ethics; intensive care unit; end-of-life; palliative
care; decision making; quality improvement

T he primary goals of intensive
care medicine are to help pa-
tients survive acute threats to
their lives while preserving
and restoring the quality of those lives.
These goals are frequently achieved, with
approximately 75% to 90% of patients
From Harvard Medical School and Children’s Hos-
pital, Boston, MA (RDT); Detroit Medical Center and
Center for Palliative Care Excellence, Wayne State
University, Detroit, MI (MLC); University of Washington,
Seattle, WA (JRC); Department of Pharmacy, Strong
Health, and Department of Surgery, School of Medicine
and Dentistry, University of Rochester, Rochester, NY
(CEH); University of California, San Francisco, CA
(JML); Division of Pulmonary and Critical Care Medi-
cine, Harborview Medical Center, University of Wash-
ington, Seattle, WA (GDR); Harriet Lane Compassionate
Care and Berman Bioethics Institute, Johns Hopkins
University and Children’s Center, Baltimore, MD (CHR);
and University of Rochester, Rochester, NY (DCK).
The American College of Critical Care Medicine
(ACCM), which honors individuals for their achieve-
ments and contributions to multidisciplinary critical
admitted to an intensive care unit (ICU)
surviving to discharge (1). Even so, the
ICU has become a common place to die;
studies show that 22% of all deaths in the
United States now occur in or after ad-
mission to an ICU (2).
care medicine, is the consultative body of the Society
of Critical Care Medicine (SCCM), which possesses
recognized expertise in the practice of critical care.
The ACCM has developed administrative guidelines
and clinical practice parameters for the critical care
practitioner. New guidelines and practice parameters
are continually developed, and current ones are sys-
tematically reviewed and revised.
Dr. Rubenfeld has held a consultancy with VERICC.
The remaining authors have not disclosed any poten-
tial conflicts of interest.
For information regarding this article, E-mail:
robert_truog@hms.harvard.edu
Copyright © 2008 by the Society of Critical Care
Medicine and Lippincott Williams & Wilkins
DOI: 10.1097/CCM.0B013E3181659096
Admission to the ICU is therefore of-
ten a therapeutic trial. Only when the
trial fails do patients and families con-
sider a change in goals, from restorative
care to palliative care. This change, which
has been called the transition from cure
to comfort, is one of the most difficult
and important aspects of medical and
nursing practice in the ICU (3). Two
truths ensure that this transition will re-
main difficult, despite our best efforts.
“First is the widespread and deeply held
desire not to be dead. Second is medi-
cine’s inability to predict the future, and
to give patients a precise, reliable prognosis
about when death will come. If death is the
alternative, many patients who have only a
small amount of hope will pay a high price
to continue the struggle” (4).
The purpose of these recommenda-
tions is to improve the care of patients
during this transition and through the
dying process. These recommendations

Crit Care Med 2008 Vol. 36, No. 3 953

build on those published in 2001 (5) and
highlight recent developments in the
field from a U.S. perspective. The recom-
mendations do not quantitatively grade
the level of evidence because most of the
recommendations are based on ethical
and legal principles that are not derived
from empirically based evidence.
Patient and Family-Centered
Care and Decision Making
Family-centered care, which sees pa-
tients as embedded within a social struc-
ture and web of relationships, is emerging
as a comprehensive ideal for end-of-life care
in the ICU (6, 7). This approach has impor-
tant implications for decision making and
communication.
Standards for Decision Making. In the
United States and many other countries,
limiting life support is ethically and le-
gally justified under the principle of au-
tonomy. U.S. law grants patients with de-
cision-making capacity the right to refuse
any and all therapies, including those that
sustain life (1). This standard is problematic
in the ICU, however, where as many as 95%
of patients may not be able to make deci-
sions for themselves because of either their
illness or sedation (8).
When patients cannot make decisions
for themselves, decisions are made on their
behalf by surrogates, using either the “sub-
stituted judgment standard” (if the pa-
tient’s values and preferences are known)
Table 1. Important U.S. court cases addressing decision making at the end of life
Case
Schloendorff v. Society of New York Hospitals, Established that competent patients have a right to
211 N.Y. 125, 105 N.E. 92 (1914)
In re Quinlan, 755 A2A 647 (H.J.), cert.
Denied, 429 U.S. 922 (1976)
Barber v. Superior Court, 147 Cal. App.
7d 1006 (Cal. App. 1983)
In re Conroy, 486A 2d 1209 (N.J., 1985)
Cruzan v. Director, Missouri Department of
Health, 497 U.S. 261 (1990)
954
or the “best interests standard” (if they are
not). While these decisions are often
reached by consensus with the patient
and family, patients do have an opportu-
nity to designate a specific individual as a
healthcare proxy. When no individual has
been specifically designated, many states
define a legal hierarchy for choosing a
designated surrogate (1). Table 1 provides
some of the legal precedents for these
principles in American law.
Legal guidelines regarding end-of-life
decision making are less clear when pa-
tients without capacity lack an appropri-
ate surrogate. Some states allow physi-
cians to make decisions for such patients
based on wishes expressed to the physi-
cians when the patients had capacity.
However, no state explicitly allows physi-
cians to make decisions based on their
view of the best interests of the patient
(1). In general, we recommend against ad
hoc decision making in these circum-
stances. One option is to ask the court to
appoint a guardian for the patient. An-
other option is to develop a clear proce-
dural guideline, including safeguards to
protect the patient’s interests (such as
mandatory ethics committee review) (9).
In an effort to engage the patient di-
rectly in decision making, there may be
times when sedatives and analgesics could
be discontinued so that a patient may re-
gain decisional capacity. Nevertheless, such
attempts may not result in a return to lu-
cidity because the patients’ illness is too
Importance
determine how their bodies can be used and that
informed consent is required before therapeutic
interventions can be performed.
Declared that competent patients have a right to
refuse interventions that, if they become
incompetent, can be exercised by surrogates
under the principle of substituted judgment.
Clarified that surrogates can refuse any and all
interventions on behalf of patients, based on a
benefit-burden analysis.
Established that, lacking surrogate knowledge of
patient wishes, decisions can be made using a
best interests standard if the burdens of
interventions outweigh their benefits and if the
pain of living is such that administering the
interventions is inhumane.
Accepted the right of competent patients to refuse
interventions but allowed states to set the level
of evidence required to determine the prior
wishes of incompetent patients with which
surrogate decisions are made.
severe. Furthermore, the attempts may be
inappropriate when the patients’ wishes are
known, death is imminent, or discontinu-
ing drugs would cause significant pain and
suffering. In these situations, surrogates
should be entrusted to make decisions for
the patient (10).
Patients and families must be given suf-
ficient time to reach decisions at the end of
life, and information should be delivered in
ways that are sensitive to the patient’s cul-
tural, religious, and language needs. Physi-
cians should take seriously their responsi-
bility to make recommendations and guide
families in ways that accord with their de-
cision-making preferences. Merely provid-
ing treatment alternatives and asking pa-
tients and families to choose among them
may make the patients and families feel
solely responsible for the decision to forgo
life-sustaining treatment, and this practice
contrasts with the preferred practice of
shared decision making (11). Some pa-
tients and families prefer to have physicians
make these decisions (12). Asking patients
and families how they prefer to make deci-
sions is an important aspect of treating
them with respect (13).
Resolving Conflict. End-of-life deci-
sions are made readily and by consensus
when communication among patients,
their surrogates, and clinicians is satis-
factory. Nevertheless, communication
may not resolve all differences, especially
when patients or families insist on inter-
ventions that clinicians consider inadvis-
able. In cases of conflict, the first step is
for all parties to focus on obtaining clar-
ity about the goals of care. Patients or
their surrogates should determine what
they hope to achieve—for example, re-
storing health, extending life, or relieving
pain and suffering. For their part, physi-
cians should provide information about
the patient’s prognosis and what goals
can be accomplished by the use of specific
interventions (14).
In the small number of cases where
agreement is not possible, physicians are
not obligated to offer therapies that they
believe cannot achieve the goals of care,
as mutually defined by the clinicians, pa-
tient, and surrogates (15, 16). When this
occurs, the American Medical Association
(17) and others have recommended a pro-
cess-based approach to resolving conflicts
over end-of-life issues. Central to this
process is mediation by hospital ethics
committees. Many hospitals have adopted
policies based on this model, and at least
one state has incorporated this approach
into legislation (18).
Crit Care Med 2008 Vol. 36, No. 3
 Communication With Families. Car- ity when clinicians spend more time dur- should possess fundamental skills in spir-
ing for family members is an important ing family conferences listening to fami- itual assessment and referral (32).
part of caring for the critically ill patient. lies, value the family input, and support
Family-centered care is based on the val- their emotions (22, 24, 26). Families also Ethical Principles Relating to the
ues, goals, and needs of the patient and need ongoing and direct communication Withdrawal of Life-Sustaining
family, including their understanding of concerning the patient’s prognosis (27, Treatment
the illness, prognosis, and treatment op- 28). A recent randomized trial showed
tions and their expectations and prefer- that such strategies to improve commu- Three ethical principles help to shape
ences for treatment and decision making nication with families can significantly the current U.S. consensus around the
(6). Supporting families through the reduce psychological morbidity in family withdrawal of life-sustaining treatment.
death of a loved one in the ICU frequently members after the ICU stay (22). Other While not all clinicians personally agree
involves guiding them through “hoping strategies for improving end-of-life com- with each of them, these principles have
for the best and planning for the worst” munication are shown in Table 2. broad-based support within the U.S. legal
(14). Compassion requires honoring the In addition, some families may choose system and accepted clinical practice and
family’s hopes and simultaneously pre- to be present during resuscitation and thereby form the basis for the specific
paring them for the possibility of death other invasive procedures (29). Research recommendations that follow (33). The
(19). Framing discussions in these terms shows that these experiences may be three principles are as follows: 1) With-
can help avoid and resolve conflicts as the highly valued by family members and holding and withdrawing life support are
patient’s condition unfolds. need not interfere with the delivery of equivalent; 2) there is an important dis-
Communication between members of medical care. Allowing family members tinction between killing and allowing to
the ICU team and the family occurs in die; and 3) the doctrine of “double effect”
to be present for such procedures re-
many settings, including the formal fam- provides an ethical rationale for provid-
quires careful planning, including guide-
ily conference as well as bedside commu- ing relief of pain and other symptoms
lines, the availability of support systems,
nication. Communication skills are an with sedatives even when this may have
and trained personnel (30).
important component of high-quality the foreseen (but not intended) conse-
Spirituality plays an important role in
critical care, and there is increasing evi- quence of hastening death (34 –37).
how some critically ill patients and clini-
dence regarding the importance of this Withholding vs. Withdrawing. Numer-
skill for family outcomes (20 –23). There cians cope with illness and death (31). ous surveys consistently show that clini-
is also increasing evidence supporting Spirituality is not synonymous with reli- cians are psychologically more comfortable
specific approaches that can improve gion. Each person’s understanding of withholding treatments than withdrawing
communication and family experiences spirituality should be explored. Assess- them (38 – 40). The reasons for this are
(22, 24, 25). Evidence shows that families ment of spiritual needs is not the exclu- complex but relate to the fact that with-
are more satisfied with communication sive domain of the chaplain but is part of holding is passive, whereas withdrawing is
and have reduced psychological morbid- the role of critical care clinicians, who active and associated with a greater sense of
moral responsibility. Despite this psycho-
logical preference, both philosophical and
Table 2. Strategies for improving end-of-life communication in the intensive care unit (ICU) legal analyses have emphasized that clini-
cians should make no distinction between
1. Communication skills training for clinicians decisions to withhold or to withdraw (41).
2. ICU family conference early in ICU course (118)
This is because whether any therapy is ini-
Evidence-based recommendations for conducting family conference:
Find a private location (21).
tiated or continued should be based solely
Increase proportion of time spent listening to family (26). on an assessment of its benefits vs. burdens
Use “VALUE” mnemonic during family conferences (22). and the preferences of the patient. Further-
Value statements made by family members. more, in many cases the value of an inter-
Acknowledge emotions. vention can only be determined after a trial
Listen to family members.
Understand who the patient is as a person. of therapy. If clinicians are reluctant to
Elicit questions from family members. withdraw therapies, they may be less in-
Identify commonly missed opportunities (25, 149). clined to give patients a trial of an indicated
Listen and respond to family members. therapy and thereby may make premature
Acknowledge and address family emotions.
Explore and focus on patient values and treatment preferences.
decisions to withhold therapies that the pa-
Affirm nonabandonment of patient and family. tient would consider beneficial.
Assure family that the patient will not suffer (24). Killing vs. Allowing to Die. In the
Provide explicit support for decisions made by the family (24). landmark case of Karen Ann Quinlan,
Additional expert opinion recommendations for conducting family conference: physicians argued that withdrawal of me-
Advance planning for the discussion among the clinical team chanical ventilation from a patient was
Identify family and clinician participants who should be involved.
Focus on the goals and values of the patient.
unethical, because it would kill the pa-
Use an open, flexible process. tient. Some philosophers have supported
Anticipate possible issues and outcomes of the discussion. this view, arguing that since killing is
Give families support and time. defined as an act that is the proximate
3. Interdisciplinary team rounds cause of a death, then withdrawal of life
4. Availability of palliative care and/or ethics consultation (115, 116)
5. Development of a supportive ICU culture for ethical practice and communication (108) support is indeed an act of killing, but
one that may be justified by the clinical

Crit Care Med 2008 Vol. 36, No. 3 955

circumstances and the consent of the pa-
tient or surrogate (33). The U.S. courts,
however, have rejected this view. In
Quinlan, for example, the court affirmed
that patients or their surrogates have the
right to refuse any unwanted medical
treatment, even if life-sustaining. There-
fore, the withdrawal of life-sustaining treat-
ments is not legally considered a killing,
and the actions of clinicians in this regard
are described as “allowing the patient to
die” from the underlying illness (42).
Intended vs. Merely Foreseen Conse-
quences. Euthanasia is illegal in the
United States, yet ICU clinicians have an
obligation to make patients comfortable
during the dying process, even when med-
ications must be administered in doses that
may shorten the patient’s life. The philo-
sophical doctrine of “double effect” is used
to draw a moral distinction between giving
medications with the intention to kill the
patient vs. giving them with the intention
to make the patient comfortable, but with
the foreseen consequence of potentially
hastening the patient’s death.
This doctrine is controversial, since it
relies on an assessment of the intentions
of the clinician, which are subjective and
can be mixed. Furthermore, in most sit-
uations individuals are morally responsi-
ble for all of the foreseen consequences of
their actions, not just those that are in-
tended (34, 36). Despite these philosoph-
ical difficulties, the principle is supported
in the U.S. law and defines the line be-
tween euthanasia and acceptable pallia-
tive care. In the words of former U.S.
Supreme Court Chief Justice Rehnquist,
“It is widely recognized that the provision
of pain medication is ethically and pro-
fessionally acceptable even when the
treatment may hasten the patient’s death
if the medication is intended to alleviate
pain and severe discomfort, not to cause
death” (43).
The intentions of the clinicians are
therefore critically important in deter-
mining the legality of the large doses of
medication that are sometimes adminis-
tered in end-of-life care. Both in verbal
communication and written documenta-
tion, clinicians must clearly express the
intention to relieve the pain and suffering
of the patient and the clinical signs that
justify the administration of additional
medications. Moreover, this intention is
evidenced by practices that rely on an
ongoing assessment of the patient’s com-
fort, coupled with judicious titration of
sedation and analgesia in accordance
with clinical guidelines
956
Some have argued that the doctrine of
double effect is not necessary, since stud-
ies suggest that the use of sedatives and
analgesics at the end of life does not ac-
tually hasten death (44 – 46). The doc-
trine is still useful, however, for justifying
those individual cases where the drugs
clearly appear to hasten death but are
necessary to control symptoms (47).
Practical Aspects of
Withdrawing Life-Sustaining
Treatments in the ICU
Practical advice around the withdrawal
of life-sustaining treatments comes from a
combination of theoretical considerations,
empirical data, and clinical experience. Al-
though the phrase “withdrawal of care” is
often heard, it is important to distinguish
between the withdrawal of life-sustaining
interventions and the withdrawal of care.
While the former is common, the latter
should never occur. Language is important,
particularly to patients and their families.
The Procedure of Withdrawing Life-
Sustaining Treatment. A useful clinical
framework for the withdrawal of life-
sustaining treatment is to consider it as
one of many critical care procedures. Cli-
nicians should follow steps familiar to
those from other procedures like tracheal
intubation or central venous catheteriza-
tion (48). Communication with the fam-
ily and preparing them for the withdrawal
process is an essential step. In particular,
families should be prepared for the nor-
mal respiratory patterns that often pre-
cede death. Clinicians should avoid refer-
ring to these as “agonal respirations,”
since this term may erroneously imply to
the family that these cause the patient to
be in “agony.” With appropriate palliative
care, this breathing pattern is a natural
part of dying and should not be associated
with any discomfort.
As with other routine clinical proce-
dures, an explicit plan of withdrawing life
support helps ensure that nothing is
overlooked, such as discontinuing rou-
tine treatments that provide no comfort
to the patient (such as chest radiographs
and blood draws). An explicit plan also
prompts busy clinicians to make impor-
tant contacts, such as with social work-
ers, clergy, and organ donation coordina-
tors (49).
Clinicians should be prepared to assist
families in the dying process. The goal is
to provide the patient and family a quiet,
private space devoid of technology and
alarms. This may be difficult in units where
curtains separate patient beds. When the
dying process is prolonged or when de-
mands for an ICU bed cannot be met in
other ways, transfer to another area in the
hospital may be unavoidable (50). The tran-
sition should occur smoothly with defer-
ence to the needs of the patient and family.
Every effort should be taken to reassure
family members that continuity of clinical
care will be maintained.
Even though excellent palliative care
can often be provided with no more than
attentive and compassionate clinical as-
sessment, there may be a tendency to
continue cardiac, pulse oximetry, and
even invasive hemodynamic monitoring
in the ICU. Since such monitoring does
not provide additional comfort to the pa-
tient and is not necessary to assess symp-
toms of distress, providers should criti-
cally review whether it should be
continued. Family members, particularly
those who have spent weeks tracking
physiologic markers, may find them-
selves paying undue attention to the
monitor instead of the patient. A specific
conversation with the family about the
rationale for stopping these forms of
monitoring may relieve anxiety.
Considerations Around Specific Ther-
apies. Once the transition from cure to
comfort has been negotiated, all ICU
therapies should be critically evaluated in
terms of whether they make a net positive
contribution to the comfort of the pa-
tient. This includes antibiotics, vasoactive
drugs, renal dialysis, and ventricular as-
sist devices. These treatments, including
intravenous fluids and nutrition, do not
provide comfort to dying patients and are
not obligatory during the withdrawal of
life support. Sometimes clinicians set lim-
its on treatments that are not currently
indicated (such as cardiopulmonary resus-
citation) while continuing other forms of
aggressive treatment. However, once the
decision has been made to withhold a life-
sustaining therapy (such as renal dialysis or
vasopressors), clinicians should critically
consider the rationale for continuing any
other life-sustaining treatments (51, 52).
There are justifications for offering
limited sets of life-sustaining treatments.
The most compelling is when a patient
has specifically refused one form of life-
sustaining treatment on the basis of per-
sonal values, as for example when a pa-
tient refuses intubation while requesting
other therapies. These wishes should be
followed as long as they are consistent
with good-quality care. However, prefer-
ences such as refusing endotracheal in-
Crit Care Med 2008 Vol. 36, No. 3
tubation while insisting on other aspects
of cardiopulmonary resuscitation are not
consistent with good-quality care and
should not be adopted as a plan of treat-
ment.
Abrupt discontinuation of life-sustain-
ing treatments, with the exception of me-
chanical ventilation, results in no discom-
fort. Therefore, there is no justification to
wean treatments such as antibiotics, blood
products, intravenous fluids, or cardiovas-
cular support. Since rapid withdrawal of
oxygen or ventilatory support may lead to
dyspnea, there is a theoretical rationale for
removing these supports gradually. Al-
though a prolonged “terminal weaning”
process has been advocated on the grounds
that it removes a sense of responsibility
from the family and clinicians, this process
is not ethically required and may, by pro-
longing the dying process, actually contrib-
ute to patient distress (53). Therefore, the
only justification for gradual reduction of
ventilatory support is to allow time to con-
trol dyspnea through the titration of med-
ications.
Considerable variation in practice at-
tends to the decision of whether to extu-
bate patients when withdrawing mechan-
ical ventilation or whether to leave the
endotracheal tube in place while the ven-
tilator is weaned (54). Since survivors of
critical illness frequently recall endotra-
cheal tubes and suctioning as significant
sources of discomfort, an argument can
be made to remove artificial airways (55).
Neither ethical principles nor current
empirical evidence can support a dog-
matic view on this question; clinicians
should solicit input from team members
and the family to make the decision in
individual cases.
The use of noninvasive ventilation
during end-of-life care should be evalu-
ated by carefully considering the goals of
care (56 –58). There are two reasonable
scenarios. A patient who has specifically
refused intubation but desires other as-
pects of intensive care with the goal of
prolonging survival may choose noninva-
sive ventilation. Alternatively, noninva-
sive ventilation may be used as a pallia-
tive technique to minimize dyspnea.
When used for the latter indication, non-
invasive ventilation should be stopped
when it is no longer effective at relieving
that symptom.
Pharmacologic Paralysis and End-of-
Life Care. Neuromuscular blocking
agents, such as pancuronium, vecuro-
nium, and atracurium, can be involved in
end-of-life care in two ways (5, 59). First,
Crit Care Med 2008 Vol. 36, No. 3
clinicians may initiate use of these agents
at the time of withdrawal of life support.
This practice, whose primary purpose is
to make the patient “look” comfortable
during the dying process, is not accept-
able. Since these agents have no sedative
or analgesic effects, their use cannot be
justified as being beneficial to the patient.
Furthermore, their use makes it impos-
sible to assess the patient’s level of com-
fort, thereby thwarting an essential goal
of end-of-life care. A second, more diffi-
cult situation occurs when a decision is
made to withdraw life support from a
patient who has been receiving paralytic
agents for therapeutic reasons. The ideal
solution in this circumstance is to allow
the paralytic agent to wear off or to phar-
macologically reverse its effects, allowing
for better assessment of the patient. In
some cases, however, the duration of ac-
tion of these agents may be very pro-
longed, as when the patient has been re-
ceiving large doses of the drugs or when
hepatic or renal failure has impaired their
clearance. In these cases, the benefits of
continuing with life support until neuro-
muscular function can be restored must be
balanced against the burdens that this sup-
port imposes on the patient and family.
The question of how to manage pa-
tients receiving paralytic agents remains
controversial. If prolonged paralysis is
recognized as an iatrogenic complication
of the patient’s treatment, then insisting
that this complication be resolved before
acceding to the families request to with-
draw life support could be seen as placing
the concerns of the care team above those
of the patient and family. These recom-
mendations therefore take the following
positions. First, paralytic agents should
never be introduced at the time of with-
drawal of life support. Second, when pa-
tients have been receiving paralytic
agents for therapeutic reasons, neuro-
muscular function should ideally be re-
stored before withdrawal of life support.
Third, when restoring neurologic func-
tion would impose an unacceptable delay
on the withdrawal of life support, with-
drawal may proceed, with particular at-
tention given to ensuring the comfort of
the patient through the dying process,
recognizing that signs of discomfort will
be difficult to detect.
Symptom Management in
End-of-Life Care
Declining or impaired cognition and
decreased consciousness are common
among ICU patients (60). When patients
cannot self-report their symptoms, clini-
cians may undertreat pain and suffering
because of difficulty in identifying behav-
ioral indicators of these symptoms (61).
Pain. More than 50% of seriously ill
hospitalized patients report some level of
pain (62). Pain in the ICU is often related
to iatrogenic causes, procedures, and in-
terventions. Moderately or severely un-
comfortable procedures that are com-
monly performed in the ICU include
suctioning, turning, catheter insertion,
wound care, and the presence of endotra-
cheal tubes (63, 64). Minimizing or elim-
inating iatrogenic sources of pain should
be part of the pain relief plan.
When patients cannot self-report their
degree of pain, standardized scoring sys-
tems based on physiologic variables and
behavioral observations can provide an
objective basis for pain management. The
Behavioral Pain Scale was developed for
use in the ICU with mechanically venti-
lated patients (65). The Behavioral Pain
Scale has strong interrater reliability,
moderate internal consistency, and dis-
criminant validity (66, 67).
The Pain Assessment Behavior Scale is
another simple, reliable, and valid instru-
ment for assessing pain in adults when a
self-report cannot be obtained. This in-
strument has strong internal consistency
and interrater reliability, and the corre-
lation between patient reports and Pain
Assessment Behavior Scale suggests that
this scale is a good measure of pain pres-
ence and intensity (68).
Dyspnea and Respiratory Distress.
Dyspnea and respiratory distress are com-
mon symptoms among patients admitted
to an ICU unit for oxygen and ventilatory
support (69). Dyspnea is defined as the
patient’s subjective awareness of altered
or uncomfortable respiratory functioning;
respiratory distress is the observable corol-
lary to dyspnea (70). Behavioral correlates
of respiratory distress in mechanically ven-
tilated patients include (in descending fre-
quency) tachypnea and tachycardia, a fear-
ful facial expression, accessory muscle use,
paradoxic breathing (diaphragmatic), and
nasal flaring (71).
The data to support specific treatment
approaches for dyspnea during end-of-life
care are sparse and incomplete. The best
approach is to individualize the treat-
ment based on the underlying source of
the dyspnea, the patient’s level of con-
sciousness, and the patient’s observed
and perceived needs. Some approaches
treat the symptom directly and thereby
957
prolong life. These include, for example, staff presence are all strategies that may
supplemental oxygen, corticosteroids, di- reduce the negative effects of delirium and
uretics, and bronchodilators. Other ap- minimize the need for sedation (76). Neu-
proaches, like administration of opioids, roleptic agents, such as haloperidol, are ef-
also make the patient comfortable but fective at reducing delirium-induced agita-
may decrease consciousness. Clinicians tion. While sedation is the hallmark
should work with patients and families to treatment for agitated delirium and is often
determine the optimal approach, or com- necessary at the end of life (10), its use
bination of approaches, for each patient deprives the patient and family of possible
on an individual basis (70). meaningful interactions before death and
Delirium. Delirium is a disturbance of should be used as a last resort.
consciousness characterized by an acute Specific Medications. The goal of drug
onset and fluctuating course such that a therapy as a component of end-of-life care
patient’s ability to receive, process, store, is the alleviation or prevention of pain, dys-
or recall information is impaired (72). pnea, and other distressing symptoms. The
When these symptoms are accompanied medications most commonly used in end-
by increased motor activity, the condition of-life care are summarized in Tables 3 and
is termed agitated delirium. Agitated de- 4. As a general rule, any time an increase in
lirium is relatively common in ICU pa- an infusion dose is being considered due to
tients as a consequence of their medical reemergence of the signs or symptoms of
condition, substance intoxication or suffering, intravenous bolus doses should
withdrawal, use of medication, or a com- be administered concurrently to achieve a
bination of these factors (60, 73), and it rapid response. The routine use of a bolus-
prompts the use of sedatives and re- infusion approach should minimize the
straints to promote patient safety and risk of unnecessary delays in response.
avoid self-harm. Delirium, calm or agi- Opioids are the mainstay for the treat-
tated, is a common symptom among con- ment of pain and dyspnea in dying pa-
scious ICU patients who are dying (74). tients and demonstrate additive sedative
Distressing symptoms, such as pain or effects as a component of drug combina-
dyspnea, may contribute to agitated delir- tions for palliative sedation (77– 81). The
ium, and analgesia and other treatments opioid analgesics recommended in recent
should be optimized before sedatives are multisociety practice guidelines are mor-
employed (75). Removing restraints, pro- phine, fentanyl, and hydromorphone
moting sleep, reducing noise and lights, (82). The effects of morphine on cardiac
and providing a soothing family member or preload may also contribute to improve-
Table 3. Opioid analgesic agents (5, 79, 82)
Onset Duration Typical
Equivalent to Peak of Effect, Typical Adult Pediatric
Dose, IVa Effect, mins hrs Dose, IV Dose, IV Typical Infusion Rate
Morphine 10 mg 20–30 3–4 2–10 mg 0.1 mg/kg 0.05–0.5 mg·kg-1·hr-1
Fentanyl 100 ␮g 2–5 0.5–2 0.5–2 ␮g/kg 1–5 ␮g/kg 0.5–10 ␮g·kg-1·hr-1
Hydromorphone 1.5–2 mg 20–30 3–4 0.5–2 mg — —
IV, intravenous.
a
Equivalent doses are approximations and are of limited value due to differences in onset and
duration of effect.
Table 4. Sedative agents (5)
Typical Initial
Onset to Peak Duration of Typical Initial Pediatric
Effect, mins Effect, hrs Adult Dose, IV Dose, IV
Sedatives
Lorazepam 20–25 2–4 1–3 mg 0.05 mg/kg
Midazolam 5–10 1.5–2 0.02–0.1 mg/kg 0.1 mg/kg
Propofol 1–2 0.1–0.4 1 mg/kg 1 mg/kg
Neuroleptics
Haloperidol 25–30 2–4 0.5–20 mg — 3–5 mg/hr
IV, intravenous.
958
ment of dyspnea in some patients by re-
ducing pulmonary edema.
Morphine is recommended as the agent
of choice for palliative care due to its effi-
cacy, low cost, familiarity to the healthcare
team, and potentially beneficial euphoric
effects. Compared with other opioids, mor-
phine is associated with a greater risk of
histamine release causing urticaria at the
injection site, pruritus, and flushing, which
may be relieved by antihistaminic therapy
(83). Fentanyl and hydromorphone are al-
ternatives to morphine. Fentanyl has a very
short duration of response and should be
administered by continuous infusion in
this setting. Although hydromorphone is
considered to have less euphoric effect than
morphine, available evidence does not sug-
gest important differences in analgesic effi-
cacy, adverse effects, or patient preference
in the management of chronic and acute
pain (84).
Benzodiazepines are the most fre-
quently used and often preferred agents
for sedation in the critical care unit, in-
cluding sedation during end-of-life care
(44, 77, 82, 85). They have no analgesic
properties; their benefits in this setting
derive from their sedative, hypnotic, an-
xiolytic, and amnestic effects. Lorazepam
and midazolam are the most commonly
used benzodiazepines. As with all cen-
trally acting drugs, the time to onset of
the benzodiazepines is primarily depen-
dent on the lipid solubility of the drug.
Midazolam is highly lipophilic and has
the most rapid onset of effect following
intravenous administration, with maxi-
mal response in approximately 5–10
mins. Lorazepam is the least lipid soluble
of the parenteral benzodiazepines and re-
quires up to 20 –25 mins to achieve max-
imal response following intravenous ad-
ministration (86, 87). There are no
convincing data of important differences
in clinical response or safety when each
agent is used appropriately.
Propofol is an intravenous general an-
Typical Initial Infusion Typical Initial Infusion
Dose, Adult Dose, Pediatric
0.5–4 mg/hr 0.05–0.1 mg·kg-1·hr-1
1–5 mg/hr 0.05–0.1 mg·kg-1·hr-1
10–50 ␮g·kg-1·min-1 10–50 ␮g·kg-1·min-1
—
Crit Care Med 2008 Vol. 36, No. 3
esthetic widely used at sedative doses in
critically ill patients. The primary advan-
tages of propofol are its very rapid onset
of effect combined with a rapid offset of
effect, allowing relatively easy titration to
the desired level of sedation (82, 88).
Other sedative agents, including the
barbiturates and ketamine, have a limited
role in this setting and can be considered
for selected patients who may be refractory
or intolerant to usual agents (47, 89).
Haloperidol is considered the drug of
choice for treatment of delirium in criti-
cally ill patients (82, 90). Due to its long
elimination half-life, haloperidol usually
requires a loading regimen for initial con-
trol of symptoms. Haloperidol reaches
maximal effect approximately 25–30 mins
following an intravenous dose and can be
repeated every 15–30 mins as needed (82).
It has been suggested that the intravenous
dose can be doubled every 30 mins until
response is achieved (91); however, single
adult doses ⬎20 mg are rarely required or
recommended. Haloperidol has no analge-
sic activity and does not have significant
sedative effects as a single agent. It is typi-
cally combined with opioid analgesics and
sedative agents to manage acute agitated
behavior in critically ill patients.
Although many drugs can be used to
treat pain and agitation at the end of life,
the importance of the practitioner’s fa-
miliarity with the drug cannot be over-
emphasized. In the last few hours of life,
there may be only one chance to prevent
pain, dyspnea, and delirium. As much ex-
pertise is necessary for the appropriate
use of drug therapy at the end-of-life as
for any other pharmacologic intervention
in critical care.
Considerations at the Time of
Death
Even when anticipated, the time sur-
rounding the death of a patient can be
stressful for both families and clinicians.
Anticipation of the tasks that must be
performed can help ensure that end-of-
life care is delivered appropriately and
compassionately.
Notification of Death. Pronouncing
death is a solemn ritual and an important
competency for end-of-life care (7). Se-
nior physician leadership and the involve-
ment of other professionals, such as the
nurse, chaplain, or social worker, are im-
portant. The communication should
avoid euphemisms and use plain lan-
guage gently and empathically (dead, dy-
ing, death, and die are all words that are
Crit Care Med 2008 Vol. 36, No. 3
rarely misunderstood) (92). Most families
need reassurance that everything appro-
priate was done to help their family mem-
ber. News of a patient’s death should be
given in person, whenever possible. When
families must be contacted by telephone,
special care should be taken in how the
information is disclosed.
Determination of Brain Death. The
frequency with which death is determined
by neurologic criteria varies greatly de-
pending on the patient population served
by an ICU, but in one study of ⬎6,000
patients who died in ICU, 6% were diag-
nosed as dead on the basis of neurologic
criteria (93). Standard criteria for ascer-
taining the diagnosis of brain death in
adults are available (94), although studies
show considerable variability in how the
diagnosis is made around the world (95).
Similarly, criteria used for determining
brain death in children are also variable,
perhaps reflecting the fact that the guide-
lines for children have not been updated
since 1987 (96). There is not general con-
sensus on the advisability of having family
members present at the time testing is per-
formed (97).
Organ Donation. Organ and tissue do-
nation is an integral part of end-of-life
decisions and bereavement practices (98,
99). Routine questioning about advance
directives and preferences at admission to
the hospital may identify potential organ
donors (100). Requests should focus on
allowing families the opportunity to de-
termine whether organ donation is con-
sistent with either the patient’s known
wishes or what the patient would likely
have wanted.
Donation by Cardiac Death (DCD) re-
quires protocols for withdrawal of life-
sustaining therapies under carefully con-
trolled conditions (101). Some have
concerns about the potential effect of
DCD on the quality of end-of-life decision
making and care in the ICU, but research
and debate are needed to develop consen-
sus regarding the best approach to
achieve high-quality palliative care si-
multaneously with allowing patients and
families the option of DCD.
Critical care professionals are respon-
sible for the integrity of the organ dona-
tion process in collaboration with the or-
gan procurement organization. Best
practices require that the request is made
in a private location and paced to give the
family time to accept the death (98).
United States regulations require that the
person requesting organ donation be spe-
cifically trained to perform this task. Re-
lational aspects of how the information is
shared with families have been shown to
be more important than details of the spe-
cific content of the information. Training
and clear role responsibilities for the inter-
disciplinary team can reduce the stress as-
sociated with requests for organ donation
and increase donation rates.
Bereavement and Support. Bereave-
ment and support services are essential to
the delivery of high-quality palliative
care, for both the family and the clini-
cians. Assisting families to cope with the
impending death, complete important
life tasks, and engage in meaningful rit-
uals is a response to the family’s antici-
patory grief. The care of the body after
death, making funeral plans, and decid-
ing about autopsy are key bereavement
tasks following a person’s death. After a
patient’s death, families benefit from in-
formation about educational and spiritual
resources, support groups, and contact
information for mental health profession-
als with expertise in bereavement (6, 7,
19, 22, 98, 102).
Similarly, clinicians also have impor-
tant bereavement needs. Although the
needs of clinicians have not been rou-
tinely addressed, their unrecognized suf-
fering and grief may undermine the ef-
fectiveness and quality of care (103, 104).
An Institute of Medicine report stressed
the importance of developing strategies
to help professionals preserve their own
integrity and well-being (105).
Needs of the Interdisciplinary Team.
The clinical team needs to be interdiscipli-
nary and committed to cooperation and
clear communication. Significant discrep-
ancies exist between critical care nurses
and physicians about satisfaction with end-
of-life decision-making processes, includ-
ing ethical issues (106, 107). Physicians
may be unaware of nurses’ perspectives on
conflict (108). More than half of the issues
identified by critical care nurses as either
obstructing or facilitating quality care for
dying patients involve some aspect of com-
munication (27).
End-of-life care requires support sys-
tems and resources for caregivers that
address moral distress, burnout, and
posttraumatic stress disorder (109 –112).
Systems of support may include regular
debriefings after patient deaths, access to
spiritual and psychosocial resources, and
relief from responsibilities for some time
after a patient dies (27, 104, 113). Within
the ICU culture, norms for appropriate
behavior, mutual support, communica-
959
tion, and resolution of ethical conflicts
are essential.
Research, Quality Improvement,
and Education
End-of-life care in the ICU, like many
aspects of critical care, offers important
opportunities for research, quality im-
provement, and education. There has been
a growing literature on each of these topics
in the past few years that can help guide
researchers, clinicians, administrators, and
educators.
Research. A recent interdisciplinary
working group identified four areas of
need for a research agenda to improve
end-of-life care: defining the problems,
identifying solutions, evaluating solutions,
and overcoming barriers. In each of these
areas, important unanswered questions
were identified (114). Two important prin-
ciples emerged: Interventions that are
likely to improve the quality of care
should be developed with preliminary
data supporting their efficacy before be-
ing subjected to randomized trials, and
outcome measures must be identified
that can be used to demonstrate the im-
provements.
One lesson from prior research in this
area is that large randomized trials of
interventions should have supporting ev-
idence from smaller studies demonstrat-
ing “proof of concept.” A number of re-
cent interventions have shown promising
results, including standardized end-of-
life family conferences with bereavement
pamphlets and routine palliative care and
ethics consultation (22, 115, 116). These
studies suggest that end-of-life care can
be improved but highlight the impor-
tance of preliminary data to establish fea-
sibility and improvement in intermediate
outcomes before launching large ran-
domized trials to provide definitive proof
of effectiveness.
The field needs to identify reliable,
valid, feasible, and responsive outcome
measures for end-of-life care. Outcome
measures used to date include length of
stay, intensity of care, families’ psycho-
logical symptoms, and ratings of quality
of care from ICU clinicians and family
members. Each of these outcomes has
important challenges. Several studies
have used as the outcome measure ICU
length of stay among patients who died
(115–118). The rationale for this out-
come measure is that if ICU days are
decreased for those patients who ulti-
mately die without an increase in overall
960
mortality rate, the intervention has re-
duced the “prolongation of dying.” There
are some important theoretical limita-
tions to using ICU days as an outcome
measure for such interventions. For ex-
ample, an intervention that rushed fam-
ilies to make decisions might be associ-
ated with decreased family satisfaction
with care and increased family depres-
sion, anxiety, or posttraumatic stress dis-
order (20, 21). Nonetheless, reducing ICU
length of stay, particularly if it is associ-
ated with high levels of family satisfaction
(115), seems like an appropriate surro-
gate marker for improved quality of care.
Another potential assessment of quality
of end-of-life care is family or clinician
satisfaction with care. Patient satisfaction
is not a practical outcome measure since
the vast majority of patients are not able
respond to questions at a time when end-
of-life care is being provided in the ICU
(119). Measures that have shown some
evidence of reliability and validity after
death in the ICU include the Quality of
Dying and Death (49, 69, 120 –122), the
Family Satisfaction with ICU (123, 124),
and the Critical Care Family Satisfaction
Survey (125). Finally, there is evidence
that improved communication about
end-of-life care can significantly reduce
psychological morbidity in family mem-
bers after a death in the ICU (22).
Quality Improvement. Many publica-
tions exist on the issue of quality im-
provement in the critical care setting
(126 –134), and there is growing recogni-
tion that end-of-life care in the ICU is an
important target for quality improvement
efforts. Recently, a working group identi-
fied the domains of quality for end-of-life
care, and these domains provide a frame-
work for understanding the diverse issues
and potential targets for quality improve-
ment efforts focused on end-of-life care in
the ICU (135). The domains were 1) pa-
tient- and family-centered decision mak-
ing; 2) communication; 3) continuity of
care; 4) emotional and practical support;
5) symptom management and comfort
care; 6) spiritual support; and 7) emo-
tional and organizational support for ICU
clinicians.
There have been several recent publi-
cations of quality improvement efforts
designed to improve end-of-life care in
the ICU. A before-after quality improve-
ment study of implementation of a stan-
dardized order form for withdrawal of
life-sustaining therapies suggests efforts
can improve care (49). There have also
been recent efforts to define process mea-
sures for high-quality end-of-life care in
the ICU (136, 137). Although these mea-
sures have not yet been validated, they
have promise for improving the quality of
care in all ICUs.
Education. Although education in this
area is improving, studies document the
deficiencies in education about end-of-
life care for physicians and nurses (138 –
144). Considerable work has been done in
education about end-of-life care that can
be adapted to the ICU setting, including
the development of training programs
such as Education for Physicians on End-
of-Life Care (145) and End-of-Life Nurs-
ing Education Consortium (146) and cur-
ricula developed for internal medicine
residency training (147). The Initiative
for Pediatric Palliative Care is a curricu-
lum focused on the care of children
(148). In addition, clinicians need oppor-
tunities to address their own emotions
and reactions to working with dying pa-
tients and their families (144). Educa-
tional programs that offer clinicians the
opportunity to talk about their experi-
ences caring for dying critically ill pa-
tients and their families and the effect of
providing this care on their own emo-
tions and work attitudes can be valuable
(113).
CONCLUSIONS
End-of-life care is emerging as a com-
prehensive area of expertise in the ICU
and demands the same high level of
knowledge and competence as all other
areas of ICU practice. There has been an
increased focus on research, education,
and quality improvement to improve end-
of-life care in the ICU setting. There is
also increasing consensus within the field
of critical care on some important prin-
ciples, such as shared decision making
and the importance of caring for patients’
families. These revised guidelines incor-
porate these recent developments in an
effort to further improve the care of pa-
tients dying in the ICU and their families.
REFERENCES
1. Luce JM, Prendergast TJ. The changing na-
ture of death in the ICU. In: Managing
Death in the Intensive Care Unit: The Tran-
sition From Cure to Comfort. Curtis JR,
Rubenfeld GD (Eds). Oxford, UK, Oxford
University Press, 2001, pp 19 –29
2. Angus DC, Barnato AE, Linde-Zwirble WT,
et al: Use of intensive care at the end of life
in the United States: An epidemiologic
study. Crit Care Med 2004; 32:638 – 643
Crit Care Med 2008 Vol. 36, No. 3
 3. Managing Death in the Intensive Care Unit:
The Transition From Cure to Comfort. Ox-
ford, UK, Oxford University Press, 2001
4. Finucane TE: How gravely ill becomes dy-
ing: A key to end-of-life care. JAMA 1999;
282:1670 –1672
5. Truog RD, Cist AF, Brackett SE, et al: Rec-
ommendations for end-of-life care in the
intensive care unit: The Ethics Committee
of the Society of Critical Care Medicine. Crit
Care Med 2001; 29:2332–2348
6. Institute of Medicine: Committee on Care at
the End of Life: Approaching Death. Wash-
ington DC, National Academy Press, 1997
7. National Consensus Project for Quality Pal-
liative Care: Clinical Practice Guidelines for
Quality Palliative Care. Pittsburgh, National
Consensus Project, 2004
8. Luce JM: Is the concept of informed con-
sent applicable to clinical research involv-
ing critically ill patients? Crit Care Med
2003; 31(3 Suppl):S153–S160
9. White DB, Curtis JR, Lo B, et al: Decisions
to limit life-sustaining treatment for criti-
cally ill patients who lack both decision-
making capacity and surrogate decision-
makers. Crit Care Med 2006; 34:2053–2059
10. Tonelli MR: Waking the dying: Must we
always attempt to involve critically ill pa-
tients in end-of-life decisions? Chest 2005;
127:637– 642
11. Heyland DK, Cook DJ, Rocker GM, et al:
Decision-making in the ICU: Perspectives of
the substitute decision-maker. Intensive
Care Med 2003; 29:75– 82
12. Azoulay E, Pochard F, Chevret S, et al: Half
the family members of intensive care unit
patients do not want to share in the deci-
sion-making process: A study in 78 French
intensive care units. Crit Care Med 2004;
32:1832–1838
13. Curtis JR: Communicating about end-of-life
care with patients and families in the inten-
sive care unit. Crit Care Clin 2004; 20:
363–380, viii
14. Back AL, Arnold RM, Quill TE: Hope for the
best, and prepare for the worst. Ann Intern
Med 2003; 138:439 – 443
15. Consensus statement of the Society of Critical
Care Medicine’s Ethics Committee regarding
futile and other possibly inadvisable treat-
ments. Crit Care Med 1997; 25:887– 891
16. Luce JM: Physicians do not have a respon-
sibility to provide futile or unreasonable
care if a patient or family insists. Crit Care
Med 1995; 23:760 –766
17. Medical futility in end-of-life care: Report of
the Council on Ethical and Judicial Affairs.
JAMA 1999; 281:937–941
18. Fine RL, Mayo TW: Resolution of futility by
due process: Early experience with the
Texas Advance Directives Act. Ann Intern
Med 2003; 138:743–746
19. Truog RD, Christ G, Browning DM, et al:
Sudden traumatic death in children: “We
did everything, but your child didn’t sur-
vive.” JAMA 2006; 295:2646 –2654
20. Azoulay E, Pochard F, Kentish-Barnes N, et
Crit Care Med 2008 Vol. 36, No. 3
al: Risk of post-traumatic stress symptoms
in family members of intensive care unit
patients. Am J Respir Crit Care Med 2005;
171:987–994
21. Pochard F, Azoulay E, Chevret S, et al:
Symptoms of anxiety and depression in
family members of intensive care unit pa-
tients: Ethical hypothesis regarding deci-
sion-making capacity. Crit Care Med 2001;
29:1893–1897
22. Lautrette A, Darmon M, Megarbane B, et al:
A communication strategy and brochure for
relatives of patients dying in the ICU.
N Engl J Med 2007; 356:469 – 478
23. Browning DM, Meyer EC, Truog RD, et al:
Difficult conversations in health care: Cul-
tivating relational learning to address the
hidden curriculum. Acad Med 2007; In
Press
24. Stapleton RD, Engelberg RA, Wenrich MD,
et al: Clinician statements and family satis-
faction with family conferences in the in-
tensive care unit. Crit Care Med 2006; 34:
1679 –1685
25. Curtis JR, Engelberg RA, Wenrich MD, et al:
Missed opportunities during family confer-
ences about end-of-life care in the intensive
care unit. Am J Respir Crit Care Med 2005;
171:844 – 849
26. McDonagh JR, Elliott TB, Engelberg RA, et
al: Family satisfaction with family confer-
ences about end-of-life care in the intensive
care unit: Increased proportion of family
speech is associated with increased satisfac-
tion. Crit Care Med 2004; 32:1484 –1488
27. Kirchhoff KT, Beckstrand RL: Critical care
nurses’ perceptions of obstacles and helpful
behaviors in providing end-of-life care to
dying patients. Am J Crit Care 2000;
9:96 –105
28. White DB, Engelberg RA, Wenrich MD, et
al: Prognostication during physician-family
discussions about limiting life support in
intensive care units. Crit Care Med 2007;
35:442– 448
29. Maclean SL, Guzzetta CE, White C, et al:
Family presence during cardiopulmonary
resuscitation and invasive procedures: Prac-
tices of critical care and emergency nurses.
Am J Crit Care 2003; 12:246 –257
30. Emergency Nurses Association: Presenting
the Option for Family Presence. Park Ridge,
IL, Emergency Nurses Association, 2001
31. Puchalski C: Spirituality in health: The role
of spirituality in critical care. Crit Care Clin
2004; 20:487–504, x
32. Robinson MR, Thiel MM, Backus MM, et al:
Matters of spirituality at the end of life in
the pediatric intensive care unit. Pediatrics
2006; 118:e719 – e729
33. Brock DW: Death and dying. In: Medical
Ethics. Veatch RM (Ed). Boston, Jones and
Bartlett, 1989, pp 329 –356
34. Quill TE, Dresser R, Brock DW: The rule of
double effect—A critique of its role in end-
of-life decision making. N Engl J Med 1997;
337:1768 –1771
35. Lo B, Rubenfeld G: Palliative sedation in
dying patients: “We turn to it when every-
thing else hasn’t worked.” JAMA 2005; 294:
1810 –1816
36. Quill TE: The ambiguity of clinical inten-
tions. N Engl J Med 1993; 329:1039 –1040
37. Sulmasy DP: Commentary: Double effect—
Intention is the solution, not the problem. J
Law Med Ethics 2000; 28:26 –29
38. Solomon MZ, Sellers DE, Heller KS, et al:
New and lingering controversies in pediat-
ric end-of-life care. Pediatrics 2005; 116:
872– 883
39. Solomon MZ, O’Donnell L, Jennings B, et
al: Decisions near the end of life: Profes-
sional views on life-sustaining treatments.
Am J Public Health 1993; 83:14 –23
40. Burns JP, Mitchell C, Griffith JL, et al: End-
of-life care in the pediatric intensive care
unit: Attitudes and practices of pediatric
critical care physicians and nurses. Crit
Care Med 2001; 29:658 – 664
41. Meisel A: Legal myths about terminating
life support. Arch Intern Med 1991; 151:
1497–1502
42. Truog RD, Levine DZ, Hutchinson T, et al:
Discontinuing immunosuppression in a
child with a renal transplant: Are there lim-
its to withdrawing life support? Discussion.
Am J Kidney Dis 2001; 38:901–915
43. Vacco v. Quill 117 S. Ct 2293. 1997
44. Sykes N, Thorns A: Sedative use in the last
week of life and the implications for end-of-
life decision making. Arch Intern Med 2003;
163:341–344
45. Sykes N, Thorns A: The use of opioids and
sedatives at the end of life. Lancet Oncol
2003; 4:312–318
46. Chan JD, Treece PD, Engelberg RA, et al:
Narcotic and benzodiazepine use after with-
drawal of life support: Association with time
to death? Chest 2004; 126:286 –293
47. Truog RD, Berde CB, Mitchell C, et al: Bar-
biturates in the care of the terminally ill.
N Engl J Med 1992; 327:1678 –1682
48. Rubenfeld GD, Crawford SW: Principles and
practice of withdrawing life sustaining
treatment in the intensive care unit. In:
Managing Death in the ICU: The Transition
from Cure to Comfort. Rubenfeld GD (Ed).
New York, Oxford University Press, 2000, pp
127–148
49. Treece PD, Engelberg RA, Crowley L, et al:
Evaluation of a standardized order form for
the withdrawal of life support in the inten-
sive care unit. Crit Care Med 2004; 32:
1141–1148
50. Society of Critical Care Medicine Ethics
Committee: Consensus statement on the
triage of critically ill patients. JAMA 1994;
271:1200 –1203
51. Asch DA, Faber-Langendoen K, Shea JA, et
al: The sequence of withdrawing life-
sustaining treatment from patients.
Am J Med 1999; 107:153–156
52. Cook D, Rocker G, Marshall J, et al: With-
drawal of mechanical ventilation in antici-
pation of death in the intensive care unit.
N Engl J Med 2003; 349:1123–1132
961
53. Gilligan T, Raffin TA: Withdrawing life sup-
port: Extubation and prolonged terminal
weans are inappropriate. Crit Care Med
1996; 24:352–353
54. Willms DC, Brewer JA: Survey of respiratory
therapists’ attitudes and concerns regarding
terminal extubation. Respir Care 2005; 50:
1046 –1049
55. Turner JS, Briggs SJ, Springhorn HE, et al:
Patients’ recollection of intensive care unit
experience. Crit Care Med 1990; 18:966 –968
56. Benditt JO: Noninvasive ventilation at the
end of life. Respir Care 2000; 45:1376 –1381
57. Levy M, Tanios MA, Nelson D, et al: Out-
comes of patients with do-not-intubate or-
ders treated with noninvasive ventilation.
Crit Care Med 2004; 32:2002–2007
58. Curtis JR, Cook DJ, Sinuff T, et al: Nonin-
vasive positive pressure ventilation in criti-
cal and palliative care settings: Understand-
ing the goals of therapy. Crit Care Med
2007; 35:932–939
59. Truog RD, Burns JP, Mitchell C, et al: Phar-
macologic paralysis and withdrawal of me-
chanical ventilation at the end of life.
N Engl J Med 2000; 342:508 –511
60. Kress JP, Hall JB: Delirium and sedation.
Crit Care Clin 2004; 20:419 – 433, ix
61. Hall P, Schroder C, Weaver L: The last 48
hours of life in long-term care: A focused
chart audit. J Am Geriatr Soc 2002; 50:
501–506
62. Desbiens NA, Mueller-Rizner N, Connors AF
Jr, et al: The symptom burden of seriously
ill hospitalized patients. SUPPORT Investi-
gators: Study to Understand Prognoses and
Preferences for Outcome and Risks of Treat-
ment. J Pain Symptom Manage 1999; 17:
248 –255
63. Nelson JE, Meier DE, Oei EJ, et al: Self-
reported symptom experience of critically
ill cancer patients receiving intensive care.
Crit Care Med 2001; 29:277–282
64. Puntillo KA, Morris AB, Thompson CL, et
al: Pain behaviors observed during six com-
mon procedures: Results from Thunder
Project II. Crit Care Med 2004; 32:421– 427
65. Payen JF, Bru O, Bosson JL, et al: Assessing
pain in critically ill sedated patients by us-
ing a behavioral pain scale. Crit Care Med
2001; 29:2258 –2263
66. Aissaoui Y, Zeggwagh AA, Zekraoui A, et al:
Validation of a behavioral pain scale in crit-
ically ill, sedated, and mechanically venti-
lated patients. Anesth Analg 2005; 101:
1470 –1476
67. Young J, Siffleet J, Nikoletti S, et al: Use of
a Behavioural Pain Scale to assess pain in
ventilated, unconscious and/or sedated pa-
tients. Intensive Crit Care Nurs 2006; 22:
32–39
68. Campbell ML, Renaud E, Vanni L: Psycho-
metric testing of a Pain Assessment Behav-
ior Scale. Paper presented at: Midwest Nurs-
ing Research Society, Cincinnati, OH,
March 2005
69. Mularski RA, Heine CE, Osborne ML, et al:
962
Quality of dying in the ICU: Ratings by
family members. Chest 2005; 128:280 –287
70. Campbell ML: Terminal dyspnea and respi-
ratory distress. Crit Care Clin 2004; 20:
403– 417
71. Campbell ML: Fear and pulmonary stress
behaviors to an asphyxial threat across cog-
nitive states. Res Nurs Health 2007; 30:
572–583
72. American Psychiatric Association: Diagnos-
tic Statistical Manual of Mental Disorders
(Report No. 4). Washington, DC: American
Psychiatric Association, 2000
73. Ely EW, Siegel MD, Inouye SK: Delirium in
the intensive care unit: An under-recog-
nized syndrome of organ dysfunction. Se-
min Respir Crit Care Med 2001; 22:115–126
74. Ely EW, Inouye SK, Bernard GR, et al: De-
lirium in mechanically ventilated patients:
Validity and reliability of the confusion as-
sessment method for the intensive care unit
(CAM-ICU). JAMA 2001; 286:2703–2710
75. Brody H, Campbell ML, Faber-Langendoen
K, et al: Withdrawing intensive life-sustain-
ing treatment—Recommendations for
compassionate clinical management.
N Engl J Med 1997; 336:652– 657
76. Harvey MA: Managing agitation in critically
ill patients. Am J Crit Care 1996; 5:7–16
77. Hawryluck LA, Harvey WR, Lemieux-
Charles L, et al: Consensus guidelines on
analgesia and sedation in dying intensive
care unit patients. BMC Med Ethics 2002;
3:E3
78. LeGrand SB, Khawam EA, Walsh D, et al:
Opioids, respiratory function, and dyspnea.
Am J Hosp Palliat Care 2003; 20:57– 61
79. Mularski RA: Pain management in the in-
tensive care unit. Crit Care Clin 2004; 20:
381– 401
80. Thomas JR, Von Gunten CF: Clinical man-
agement of dyspnoea. Lancet Oncol 2002;
3:223–228
81. Jennings AL, Davies AN, Higgins JP, et al: A
systematic review of the use of opioids in
the management of dyspnoea. Thorax 2002;
57:939 –944
82. Jacobi J, Fraser GL, Coursin DB, et al: Clin-
ical practice guidelines for the sustained
use of sedatives and analgesics in the criti-
cally ill adult. Crit Care Med 2002; 30:
119 –141
83. Balestrieri F, Fisher S: Analgesics. In: The
Pharmacologic Approach to the Critically Ill
Patient. 3rd ed. Chernow B (Ed). Baltimore:
Williams & Wilkins, 1994, pp 640 – 650
84. Quigley C, Wiffen P: A systematic review of
hydromorphone in acute and chronic pain.
J Pain Symptom Manage 2003; 25:169 –178
85. Cowan JD, Walsh D: Terminal sedation in
palliative medicine—Definition and review
of the literature. Support Care Cancer
2001; 9:403– 407
86. Charney DS, Mihic SJ, Harris RA: Hypnotics
and sedatives. In: The Pharmacologic Basis
of Therapeutics. 11th ed. Brunton LL, Lazo
JS, Parker KL (Eds). New York, McGraw-
Hill, 2006, pp 401– 427
87. Greenblatt DJ. Sedation: Intravenous ben-
zodiazepines in critical care medicine. In:
The Pharmacologic Approach to the Criti-
cally Ill Patient. 3rd ed. Chernow B (Ed).
Baltimore, Williams & Wilkins, 1994, pp
321–326
88. Evers AS, Crowder CM. General anesthetics.
In: Goodman & Gilman’s The Pharmaco-
logic Basis of Therapeutics. 11th ed. Brun-
ton LL, Lazo JS, Parker KL (Eds). New York,
McGraw-Hill, 2006, pp 341–368
89. Berger JM, Ryan A, Vadivelu N, et al: Ket-
amine-fentanyl-midazolam infusion for
the control of symptoms in terminal life
care. Am J Hosp Palliat Care 2000; 17:
127–134
90. Ely EW, Stephens RK, Jackson JC, et al:
Current opinions regarding the impor-
tance, diagnosis, and management of delir-
ium in the intensive care unit: A survey of
912 healthcare professionals. Crit Care Med
2004; 32:106 –112
91. Fernandez F, Adams F, Levy JK, et al: Cog-
nitive impairment due to AIDS-related
complex and its response to psychostimu-
lants. Psychosomatics 1988; 29:38 – 46
92. Hallenbeck J: Palliative care in the final
days of life: “They were expecting it at any
time.” JAMA 2005; 293:2265–2271
93. Prendergast TJ, Claessens MT, Luce JM: A
national survey of end-of-life care for criti-
cally ill patients. Am J Respir Crit Care Med
1998; 158:1163–1167
94. Wijdicks EFM: The diagnosis of brain death.
N Engl J Med 2001; 344:1215–1221
95. Wijdicks EFM: Brain death worldwide—
Accepted fact but no global consensus in
diagnostic criteria. Neurology 2002; 58:
20 –25
96. Report of Special Task Force. Guidelines for
the determination of brain death in chil-
dren. Pediatrics 1987; 80:298 –300
97. Ropper AH: Unusual spontaneous move-
ments in brain-dead patients. Neurology
1984; 34:1089 –1092
98. Williams MA, Lipsett PA, Rushton CH, et al:
The physician’s role in discussing organ do-
nation with families. Crit Care Med 2003;
31:1568 –1573
99. Institute of Medicine: Organ Donation: Op-
portunities for Action. Washington, DC,
The National Academies Press, 2006
100. Essebag V, Cantarovich M, Crelinsten G:
Routine advance directive and organ dona-
tion questioning on admission to hospital.
Ann R Coll Physicians Surg Can 2002; 35:
225–231
101. Institute of Medicine: Non-Heart-Beating
Organ Transplantation: Practice and Proto-
cols. Washington, DC, National Academy
Press, 2000
102. Meyer EC, Burns JP, Griffith JL, et al: Pa-
rental perspectives on end-of-life care in the
pediatric intensive care unit. Crit Care Med
2002; 30:226 –231
103. Rushton CH: Care-giver suffering in critical
care nursing. Heart Lung 1992; 21:303–306
104. Rushton CH, Reder E, Hall B, et al: Inter-
Crit Care Med 2008 Vol. 36, No. 3
 disciplinary interventions to improve pedi-
atric palliative care and reduce health care
professional suffering. J Palliat Med 2006;
9:922–933
105. Institute of Medicine: When Children Die:
Improving Palliative and End-of-Life Care
for Children and Their Families. Washing-
ton, DC, National Academy Press, 2003
106. Ferrand E, Lemaire F, Regnier B, et al:
Discrepancies between perceptions by phy-
sicians and nursing staff of intensive care
unit end-of-life decisions. Am J Respir Crit
Care Med 2003; 167:1310 –1315
107. Oberle K, Hughes D: Doctors’ and nurses’
perceptions of ethical problems in end-of-
life decisions. J Adv Nurs 2001; 33:707–715
108. Hamric AB, Blackhall LJ: Nurse-physician
perspectives on the care of dying patients in
intensive care units: Collaboration, moral
distress, and ethical climate. Crit Care Med
2007; 35:422– 429
109. American Association of Critical Care Nurs-
es: Moral Distress Position Statement. Ali-
sio Viego, CA, Association of Critical Care
Nurses, January 12, 2005
110. Embriaco N, Azoulay E, Barrau K, et al:
High level of burnout in intensivists: Prev-
alence and associated factors. Am J Respir
Crit Care Med 2007; 175:686 – 692
111. Poncet MC, Toullic P, Papazian L, et al:
Burnout syndrome in critical care nursing
staff. Am J Respir Crit Care Med 2006; 175:
698 –704
112. Mealer ML, Shelton A, Berg B, et al: In-
creased prevalence of post traumatic stress
disorder symptoms in critical care nurses.
Am J Respir Crit Care Med 2006; 175:
693– 697
113. Hough CL, Hudson LD, Salud A, et al:
Death rounds: End-of-life discussions
among medical residents in the intensive
care unit. J Crit Care 2005; 20:20 –25
114. Rubenfeld GD, Curtis JR: End-of-life care in
the intensive care unit: A research agenda.
Crit Care Med 2001; 29:2001–2006
115. Schneiderman LJ, Gilmer T, Teetzel HD, et
al: Effect of ethics consultations on nonben-
eficial life-sustaining treatments in the in-
tensive care setting: A randomized con-
trolled trial. JAMA 2003; 290:1166 –1172
116. Campbell ML, Guzman JA: Impact of a pro-
active approach to improve end-of-life care
in a medical ICU. Chest 2003; 123:266 –271
117. Campbell ML, Guzman JA: A proactive ap-
proach to improve end-of-life care in a med-
ical intensive care unit for patients with
terminal dementia. Crit Care Med 2004; 32:
1839 –1843
118. Lilly CM, De Meo DL, Sonna LA, et al: An
intensive communication intervention for
the critically ill. Am J Med 2000; 109:
469 – 475
119. Prendergast TJ, Luce JM: Increasing inci-
Crit Care Med 2008 Vol. 36, No. 3
dence of withholding and withdrawal of life
support from the critically ill. Am J Respir
Crit Care Med 1997; 155:15–20
120. Levy CR, Ely EW, Payne K, et al: Quality of
dying and death in two medical ICUs: Per-
ceptions of family and clinicians. Chest
2005; 127:1775–1183
121. Hodde NM, Engelberg RA, Treece PD, et al:
Factors associated with nurse assessment of
the quality of dying and death in the inten-
sive care unit. Crit Care Med 2004; 32:
1648 –1653
122. Curtis JR, Patrick DL, Engelberg RA, et al: A
measure of the quality of dying and death:
Initial validation using after-death inter-
views with family members. J Pain Symp-
tom Manage 2002; 24:17–31
123. Heyland DK, Rocker GM, Dodek PM, et al:
Family satisfaction with care in the inten-
sive care unit: Results of a multiple center
study. Crit Care Med 2002; 30:1413–1418
124. Heyland DK, Tranmer JE: Measuring family
satisfaction with care in the intensive care
unit: Development of a questionnaire and
preliminary data. Crit Care Med 2002; 16:
142–149
125. Wasser T, Matchett S: Final version of the
Critical Care Family Satisfaction Survey
questionnaire. Crit Care Med 2001; 29:
1654 –1655
126. Frutiger A, Moreno R, Thijs L, et al: A cli-
nician’s guide to the use of quality termi-
nology. Working Group on Quality Im-
provement of the European Society of
Intensive Care Medicine. Intensive Care
Med 1998; 24:860 – 863
127. Angus DC, Black N: Improving care of the
critically ill: Institutional and health-care
system approaches. Lancet 2004; 363:
1314 –1320
128. Bion JF, Heffner JE: Challenges in the care
of the acutely ill. Lancet 2004; 363:970 –977
129. Cook DJ, Montori VM, McMullin JP, et al:
Improving patients’ safety locally: Changing
clinician behaviour. Lancet 2004; 363:
1224 –1230
130. Lilford R, Mohammed MA, Spiegelhalter D,
et al: Use and misuse of process and out-
come data in managing performance of
acute medical care: Avoiding institutional
stigma. Lancet 2004; 363:1147–1154
131. Pronovost PJ, Nolan T, Zeger S, et al: How
can clinicians measure safety and quality in
acute care? Lancet 2004; 363:1061–1067
132. Garland A: Improving the ICU: Part 1. Chest
2005; 127:2151–2164
133. Garland A: Improving the ICU: Part 2. Chest
2005; 127:2165–2179
134. Curtis JR, Cook DJ, Wall RJ, et al: Intensive
care unit quality improvement: A “how-to”
guide for the interdisciplinary team. Crit
Care Med 2006; 34:211–218
135. Clarke EB, Curtis JR, Luce JM, et al: Quality
indicators for end-of-life care in the inten-
sive care unit. Crit Care Med 2003; 31:
2255–2262
136. Nelson JE, Mulkerin CM, Adams LL, et al:
Improving comfort and communication in
the ICU: A practical new tool for palliative
care performance measurement and feed-
back. Qual Saf Health Care 2006; 15:
264 –271
137. Mularski RA, Curtis JR, Billings JA: Pro-
posed quality measures for palliative care in
the critically ill: A consensus from the Rob-
ert Wood Johnson Foundation Critical Care
Workgroup. Crit Care Med 2006; 34(11
Suppl):S404 –S411
138. Billings JA, Block S: Palliative care in un-
dergraduate medical education: Status re-
port and future directions. JAMA 1997; 278:
733–738
139. Weissman DE, Block SD, Blank L, et al:
Recommendations for incorporating pallia-
tive care education into the acute care hos-
pital setting. Acad Med 1999; 74:871– 877
140. Sullivan AM, Lakoma MD, Block SD: The
status of medical education in end-of-life
care: A national report. J Gen Intern Med
2003; 18:685– 695
141. Sullivan AM, Warren AG, Lakoma MD, et al:
End-of-life care in the curriculum: A na-
tional study of medical education deans.
Acad Med 2004; 79:760 –768
142. Block SD: Medical education in end-of-life
care: The status of reform. J Palliat Med
2002; 5:243–248
143. Ferrell BR, Grant M: Nurses cannot practice
what they do not know. J Prof Nurs 2001;
17:107–108
144. Redinbaugh EM, Sullivan AM, Block SD, et
al: Doctors’ emotional reactions to recent
death of a patient: Cross sectional study of
hospital doctors. BMJ 2003; 327:185
145. Ferris FD, Von Gunten CF, Emanuel LL:
Competency in end-of-life care: Last hours
of life. J Palliat Med 2003; 6:605– 613
146. Matzo ML, Sherman DW, Penn B, et al: The
end-of-life nursing education consortium
(ELNEC) experience. Nurse Educ 2003; 28:
266 –270
147. Weissman DE, Mullan PB, Ambuel B, et al:
End-of-life curriculum reform: Outcomes
and impact in a follow-up study of internal
medicine residency programs. J Palliat Med
2002; 5:497–506
148. Browning DM, Solomon MZ: The initiative
for pediatric palliative care: An interdisci-
plinary educational approach for healthcare
professionals. J Pediatr Nurs 2005; 20:
326 –334
149. West HF, Engelberg RA, Wenrich MD, et al:
Expressions of nonabandonment during the
intensive care unit family conference. J Pal-
liat Med 2005; 8:797– 807
963